Manufacturing Fellowship Extension in: Controlling geometrical variability of products in the digital and smart manufacturing era
Lead Research Organisation:
University of Huddersfield
Department Name: Sch of Computing and Engineering
Abstract
This fellowship proposal is a three year extension of the current EPSRC manufacturing fellowship: Controlling Geometrical Variability of Products for Manufacturing (EP/I033424/1). The current fellowship is exploring the mathematical fundaments for the decomposition of geometry (i.e. size, shape and texture) and creating ground-breaking technology to control geometrical variability in manufactured products. The approach links fundamental geometrical mathematics direct to key component's design, manufacturing and verification from different industrial sectors (i.e. aerospace, optics, healthcare and catapult centres). In this case, the different types of geometrical decompositions specified geometrical surface requirements (spectrum, morphological and segmentation decompositions).
The fellowship extension proposal will take the research results from the current fellowship and use them as a stepping stone for more advanced fundamental research in new areas within the manufacturing value chain. The research work is broken down to four aspects:
1. Different aspects of the manufacturing process leave different multi-scalar geometrical features, in a surface, at different scales (i.e. size, shape and texture). By decomposing these different signature features, information regarding different manufacturing aspects can be gained enabling characterisation and control of different aspects of the manufacturing process.
2. Sensor network provide information in the form of an irregular image, like a cubist painting, with different views, and times, of the environment all combined together. Decomposition of this information will provide access to features, and their relationships enabling an agile dynamic predictive model to be self-aware of its environment enabling mathematical foundations for bio-inspired feedback control loops from sensor networks to be developed.
3. Smart autonomous manufacturing will require access to the huge amassed manufacturing knowledge-base (National and International Standards, Materials data-sheets, etc.). Create the foundations of decomposition of information structures for the automatic creation of smart information systems that are machine readable and to apply this result to develop the full rigorous mathematical foundations for the manufacturing value chain.
4. Using the EPSRC Future HUB in Advanced Metrology (EP/P006930/1) as leverage, disseminate the results from the above to solve real industrial problems to demonstrate the advantage of using fundamental decomposition theory, as developed in the previous manufacturing fellowship and this extension, over traditional approaches.
The fellowship extension proposal will take the research results from the current fellowship and use them as a stepping stone for more advanced fundamental research in new areas within the manufacturing value chain. The research work is broken down to four aspects:
1. Different aspects of the manufacturing process leave different multi-scalar geometrical features, in a surface, at different scales (i.e. size, shape and texture). By decomposing these different signature features, information regarding different manufacturing aspects can be gained enabling characterisation and control of different aspects of the manufacturing process.
2. Sensor network provide information in the form of an irregular image, like a cubist painting, with different views, and times, of the environment all combined together. Decomposition of this information will provide access to features, and their relationships enabling an agile dynamic predictive model to be self-aware of its environment enabling mathematical foundations for bio-inspired feedback control loops from sensor networks to be developed.
3. Smart autonomous manufacturing will require access to the huge amassed manufacturing knowledge-base (National and International Standards, Materials data-sheets, etc.). Create the foundations of decomposition of information structures for the automatic creation of smart information systems that are machine readable and to apply this result to develop the full rigorous mathematical foundations for the manufacturing value chain.
4. Using the EPSRC Future HUB in Advanced Metrology (EP/P006930/1) as leverage, disseminate the results from the above to solve real industrial problems to demonstrate the advantage of using fundamental decomposition theory, as developed in the previous manufacturing fellowship and this extension, over traditional approaches.
Planned Impact
Academic community: Decomposition is the technique behind analysis: the process of breaking a complex topic into smaller parts in order to gain a better understanding of it. This approach is universal in academia, particularly STEM subjects. By providing a universal framework for decomposition, it is believed this will encourage cross fertilisation between many academic fields: growing knowledge and understanding in all the subject areas. Of particular importance is the decomposition of:
1) knowledge structures. The creation of smart information systems, within a particular academic field, will certainly increase the academic productivity within each field and encourage novel and enhanced teaching, from schools to PhDs.
2) bio-inspired feedback control loops from sensor networks. Control underpins most cyber-physical systems related research (e.g. smart buildings, structures and cities, next generation advanced nuclear manufacture and operations etc.). Decomposition of the sensor network information into features and patterns for control will also provide a foundation for such systems to achieve system level autonomous control.
Industry sectors: Using the EPSRC Future HUB in Advanced Metrology (EP/P006930/1) as leverage together with other HUBs and catapult centres, the research results will be disseminated to enhance the solution of real industrial problems across a range of industry sectors including: aerospace, automotive, bio-medical, electronics, energy generation, instruments & sensors, optics, and wider precision engineering. The improvements in design, manufacturing and verification capabilities will enable maximum improvements in manufacturing productivity, capability, quality and cost.
Embedded sensor networkss, metrology-driven control, and semantic systems can all be applied to other academic fields, where metrology-driven autonomous systems can also be generated allowing efficiencies in design, cost and performance
Policy makers and public corporations: The fellowship extension will also have an input into the Hub acting as a national focal point for Advanced Metrology and influence policy-makers through the national measurement strategy and manufacturing metrology standardisation.
Timescales
Short-term: By continuing the underlying research and exploring new approaches to this research, it is envisaged a new generation of decomposition-driven technologies will be delivered quickly. Collaborating companies will quickly assimilate the novel technologies, techniques and processes. Knowledge transfer activities, including metrology training, will also see significant industry benefits in skills development and decomposition best practice.
Medium term: The standardisation of the fundamental toolbox of practical advanced decomposition techniques. Within the British standards institute, the PI is convenor of committees TDW4/-/3. Here the results from the fellowship extension will be made available for national standardisation, which will help UK industry have a priority in using them. Further, it can finally propose to be a new ISO technical specification which can be published quickly for a wide range of industries to try out. Three years after publication technical specifications are reviewed and if found acceptable converted into full ISO standards.
Long-term: The potential impact of the research for the economic competitiveness of the UK is substantial and will facilitate integrated, digitalised and autonomous manufacturing. The UK will see greater productivity and efficiency, an increase in GVA contributions and a much higher skilled sector as a direct result of the extended fellowship: further developing UK manufacturing.
1) knowledge structures. The creation of smart information systems, within a particular academic field, will certainly increase the academic productivity within each field and encourage novel and enhanced teaching, from schools to PhDs.
2) bio-inspired feedback control loops from sensor networks. Control underpins most cyber-physical systems related research (e.g. smart buildings, structures and cities, next generation advanced nuclear manufacture and operations etc.). Decomposition of the sensor network information into features and patterns for control will also provide a foundation for such systems to achieve system level autonomous control.
Industry sectors: Using the EPSRC Future HUB in Advanced Metrology (EP/P006930/1) as leverage together with other HUBs and catapult centres, the research results will be disseminated to enhance the solution of real industrial problems across a range of industry sectors including: aerospace, automotive, bio-medical, electronics, energy generation, instruments & sensors, optics, and wider precision engineering. The improvements in design, manufacturing and verification capabilities will enable maximum improvements in manufacturing productivity, capability, quality and cost.
Embedded sensor networkss, metrology-driven control, and semantic systems can all be applied to other academic fields, where metrology-driven autonomous systems can also be generated allowing efficiencies in design, cost and performance
Policy makers and public corporations: The fellowship extension will also have an input into the Hub acting as a national focal point for Advanced Metrology and influence policy-makers through the national measurement strategy and manufacturing metrology standardisation.
Timescales
Short-term: By continuing the underlying research and exploring new approaches to this research, it is envisaged a new generation of decomposition-driven technologies will be delivered quickly. Collaborating companies will quickly assimilate the novel technologies, techniques and processes. Knowledge transfer activities, including metrology training, will also see significant industry benefits in skills development and decomposition best practice.
Medium term: The standardisation of the fundamental toolbox of practical advanced decomposition techniques. Within the British standards institute, the PI is convenor of committees TDW4/-/3. Here the results from the fellowship extension will be made available for national standardisation, which will help UK industry have a priority in using them. Further, it can finally propose to be a new ISO technical specification which can be published quickly for a wide range of industries to try out. Three years after publication technical specifications are reviewed and if found acceptable converted into full ISO standards.
Long-term: The potential impact of the research for the economic competitiveness of the UK is substantial and will facilitate integrated, digitalised and autonomous manufacturing. The UK will see greater productivity and efficiency, an increase in GVA contributions and a much higher skilled sector as a direct result of the extended fellowship: further developing UK manufacturing.
Organisations
- University of Huddersfield (Fellow, Lead Research Organisation)
- International Standards Organization (Collaboration)
- National Institute of Standards & Technology (NIST) (Collaboration)
- Digital Surf (Collaboration)
- Polytechnic University of Milan (Collaboration)
- OCF Plc (Project Partner)
- Royal National Orthopaedic Hospital (Project Partner)
- High Value Manufacturing Catapult (Project Partner)
- National Physical Laboratory (Project Partner)
- Ametek (United Kingdom) (Project Partner)
Publications
Anwer N
(2018)
Toward a Classification of Partitioning Operations for Standardization of Geometrical Product Specifications and Verification
in Procedia CIRP
Cai N
(2020)
A new partitioning process for geometrical product specifications and verification
in Precision Engineering
Cui W.
(2023)
An intelligent surface segmentation approach based on U-Net for structured and freeform surface characterisation
in European Society for Precision Engineering and Nanotechnology, Conference Proceedings - 23rd International Conference and Exhibition, EUSPEN 2023
Cui W.
(2023)
A novel surface denoising approach based on deep learning for freeform structured surface in metrology
in Laser Metrology and Machine Performance XV - 15th International Conference and Exhibition on Laser Metrology, Machine Tool, CMM and Robotic Performance, LAMDAMAP 2023
Huang H.
(2023)
Comparison of discrete Laplacian-Beltrami operators and their applications in surface metrology
in Laser Metrology and Machine Performance XV - 15th International Conference and Exhibition on Laser Metrology, Machine Tool, CMM and Robotic Performance, LAMDAMAP 2023
Jiang X
(2021)
Feature-based characterisation of surface topography and its application
in CIRP Annals
Jiang X
(2019)
CIRP Encyclopedia of Production Engineering
Liu W.
(2022)
INVESTIGATE GEOMETRY AND SURFACE TEXTURE OF SELECTIVE LASER MELTING BUILT INTERNAL CHANNELS WITH VARYING INCLINATION ANGLES USING XCT
in 2022 ASPE and euspen Summer Topical Meeting on Advancing Precision in Additive Manufacturing
Description | 1/ Developed surface texture analytics for triangular mesh data. Have written and currently writing software modules to implement these research results into commercial software (digital surf mountains map).. 2/ Developed a semantic hierarchical model for senor nets. Will incorporate this into CSL code. 3/ Developing Category Semantic Language (CSL). Currently being beta tested by NIST USA. Next stage develop graphical interfaces for different applications 4/ Developing Case Study with Milan Polytechnic for optimising a sensor net for a real sensor net (Large Swedish Furniture manufacturer) 5/ Developing Case Study with ISO TC/213 to restructure the Geometrical Product Standards into a semantic structure that is machine readable for future AI systems. |
Exploitation Route | 1/ Triangular mesh analytics will form the basis of future ISO standards in ISO TC/213 Geometrical Product Specification and Verification;. 2/ Sensor net model has been developed for Smart manufacturing particularly autonomous manufacturing: Case study with Milan Polytechnic; 3/ CSL is also being developed for AI ready documants for use with AI systems. This is to capture human knowledge in a semantic structured way for utility by AI systems. Case study with ISO TC213. 4/ How the latest measurement theory and technologies, underpin zero waste and carbon neutral smart manufacturing growth and productivity |
Sectors | Aerospace Defence and Marine Digital/Communication/Information Technologies (including Software) Education Energy Environment Manufacturing including Industrial Biotechology |
Description | 2022: The developed extend measurement theory and technologies are starting to underpin zero waste and carbon neutral smart manufacturing growth and productivity (i.e. what sort of new metrology theory is required for sustainability). 2024: Measurement consists of: knowledge of an unknow unobserved information structure is required. Through measurement we observe a proxy information structure which through appropriate reconstruction gives the required knowledge. In classical metrology the these information structures are single numbers and their structure a linear ordering of these numbers. What information structures does growing a sustainable future, the realisation of net zero, and a smart manufacturing driven economy require? How can we stably measure these information structures for actionable information? I have now identified the necessary and sufficient conditions on the forward measurement mappings, together with the reconstruction inverse mappings to gain stable knowledge of the unknown unobserved information structure from the observed proxy information structure. This is important in particular for taxonomies and classifications for sustainability, zero waste, carbon neutrality, smart manufacturing growth and productivity. |
First Year Of Impact | 2023 |
Sector | Aerospace, Defence and Marine,Energy,Environment,Manufacturing, including Industrial Biotechology,Transport |
Impact Types | Societal Economic Policy & public services |
Title | Abstraction and Category Semantic Language |
Description | Creative use of the Concept of Abstraction for Advanced Metrology and other research areas. The current highest level of abstraction in Mathematics is Category Theory. A Category Semantic Language has been developed to model these abstraction hierarchical structures and is a new software language for smart semantic hierarchy structure applications and uses category theory at a high semantic level. It models semantic structures where: static structure are the nouns in the language and dynamic information flows are the verbs. These are put into an hierarchical structure where each level is at the same abstraction level and higher levels are abstracted versions of lower abstraction levels. Hybrid delta lens (category theory) are used model the information flows between one abstraction level to another level. |
Type Of Material | Improvements to research infrastructure |
Year Produced | 2021 |
Provided To Others? | Yes |
Impact | Together with Milan University used these smart abstraction hierarchical structures to model a digital twin model for an assembly line, improving computational speed and reduced the number of sensors required. Together with NIST USA, this technique is being used to model the structure of the ISO GPS system (over 150 ISO documents) together with the concept of future smart ISO documents. A proposition document has been presented to ISO TC/213 in September 2022 and a special talk to the whole of ISO TC/213 on Friday 3rd May at BSI Chiswich London. I have also researched the structures of documents (paper, electronic, XPL, etc.) and proposed a new abstracted structure for all based on three pillars to be used for future smart documents. The three pillars are 1) content - text, figures, other information structures, and the flow of information between all within a document; 2) labelling the content for identification - page numbers, chapter and verse, labelling structures and flow structures; 3) quick access to content - chapter names and locations, indexes and locations, search engines, query languages, smart knowledge graphs, smart interpretation queries, etc. This is being researched for next generation 'smart documents'. This structure of documents goes well beyond traditional documents to: for example sensor nets and whole system models. |
Title | Metrology decomposition and semantic labelling |
Description | The future of metrology is decomposition: decomposition of measurement data into stable and robust features (both numerical and semantic decomposition) followed by possible semantic labelling of those feature (through their attributes and relationships between features). This will ensure smart AI systems can: read, query, and interpret the original measured data. A Key-Note presentation and paper was presented at the CIRP 2021 summer conference. |
Type Of Material | Improvements to research infrastructure |
Year Produced | 2021 |
Provided To Others? | Yes |
Impact | DigitalSurf (France) (World leading surface texture software) and Taylor Hobson (UK) (World leading surface texture instrument manufacturer) have both developed products based on the concepts developed. |
Description | Digitalsurf software |
Organisation | Digital Surf |
Country | France |
Sector | Private |
PI Contribution | Developing software modules for their mountains software package. In particular analytics on triangular mesh structures for freeform geometries. This include best fit form algorithms, filtration algorithms and characterisation parameters for freeform surface texture. All this has resulted from research developed from the EPSRC fellowships. Currently their software deals with lattice of heights on a planar surface. Have now moved to phase two to cover more software modules. |
Collaborator Contribution | They are providing a commercial framework (Mountains software package) to market our software modules. |
Impact | The first phase of this collaboration was launched in May 2020. Phase two has already begun. |
Start Year | 2019 |
Description | International Standards Organisation Technical Committee 213 - Geometrical Product Specifications and Verification |
Organisation | International Standards Organization |
Country | Switzerland |
Sector | Charity/Non Profit |
PI Contribution | Provided the mathematical hierarchical structure to model the semantic structure of the GPS system and the semantic language to describe it and the rules for stability and consistency (Category Semantic Language). Further we have 15 years experience in researching and developing such models. Presented a position document to ISO TC/213 to move the ISO GPS system forward to the smart era. Gave a public presentation to ISO TC/213 (Friday 3rd March 2023) based on the position document. |
Collaborator Contribution | Provided the domain knowledge to be incorporated into the hierarchical model and to make sure that it is still usage by humans as well as AI systems. |
Impact | ISO TC/213 held a meeting (Thursday 9th March 2023) to plan the way forward, based on the position document and public talk. |
Start Year | 2021 |
Description | Milan Polytechnic |
Organisation | Polytechnic University of Milan |
Country | Italy |
Sector | Academic/University |
PI Contribution | Providing the mathematical hierarchical structure to model the assembly line and the sensor nets it contains. |
Collaborator Contribution | Brought the original problem (to which they were asked to solve by the company with the problem) to the project. Traditional ontology approached to solve the problem did not contain a full structure to obtain a solution Also they bring domain expertise to help build the appropriate model to optimise the placing and number of sensors in the sensor net. |
Impact | Paper presented for CIRP summer 2022 meeting (via Zoom) Paper published in the proceedings of the Royal Society A, 2022 |
Start Year | 2021 |
Description | NIST Gaitherburg USA |
Organisation | National Institute of Standards & Technology (NIST) |
Country | United States |
Sector | Public |
PI Contribution | Developed a Category Semantic Language (CSL)for smart databases. This is a high level language all other category languages are at a very low level |
Collaborator Contribution | NIST have their own category query language but at a very low level. They are testing our beta version of CSL by first understanding the new concepts in CSL, second trying out case studies to test the correctness of the language and thirdly advising on suitable interfaces for CSL for different applications. |
Impact | Too early for outcomes |
Start Year | 2019 |
Title | Triangular mesh analytics software and book |
Description | New analytics for Freeform geometries using triangular mesh. This includes: 1/ best fitting freeform geometry algorithms; 2/ Filtration algorithms - PDE based filtration methods, segmentation and morphological filters; 3/ Surface texture characterisation parameters calculated from a triangular mesh |
IP Reference | |
Protection | Trade Mark |
Year Protection Granted | 2019 |
Licensed | Yes |
Impact | Phase I of the software was released in Oct 2020 to Digital Surf for their Mountains Map surface texture analytics package (the worlds leading surface texture analytics software). This takes the package to another dimension going from the analysis of planar measurement data to true 3D measured data. We have a commercial agreement between Digital surf and the University with a road-map of future research and associated software releases. Phase II of the software release is currently underway. Book entitled: "Advanced Metrology - Freeform Surfaces" has been published by Academic Press, ISBN 978-0-12-821815-0 in 2021. Co-authored by X. Jane Jiang and Paul J. Scott. This book was released in conjunction with the above software. |
Title | Triangular mesh analytics software |
Description | Software modules for Digitalsurf mountains map for surface texture analytics on trinagular meshes. |
Type Of Technology | Software |
Year Produced | 2020 |
Impact | Phase I of the software was released in Oct 2020 to Digital Surf for their Mountains Map surface texture analytics package (the worlds leading surface texture analytics software). This takes the package to another dimension going from the analysis of planar measurement data to true 3D measured data. We have a commercial agreement between Digital surf and the University with a road-map of future research and associated software releases. |
Description | International Standards |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | I take active part in ISO TC213 (geometrical product specification). I am convener of two working groups WG15 (filtration and extraction) and AG12 (mathematics for GPS). I am also an active member (UK principle expert) of WP16 (surface texture). I am a member of TC213's strategic board (AG1) and its final editing board AG2). I have been project leader of 23 ISO standards. |
Year(s) Of Engagement Activity | 2018,2019,2020,2021,2022,2023 |